An Integrative Theory of Motivation, Volition, and...

An Integrative Theory of Motivation,Volition, and Performance

JOHN KELLER*

307 Stone Building, MC 4453, Florida State University, Tallahassee, FL 32306-4453 USA

There are many independent theories and constructs pertaining to motivation,volition, and learning and the total number continues to expand. In contrast,it could be beneficial for guiding research and practice to have more theoriesthat are integrative in nature; that help explain relationships among theories inrelation to motivation, volition, learning, and performance. Typically,researchers focus on a bounded set of questions within a given area of inter-est that incorporates a specific paradigm of inquiry. However, to have inte-grative theories it is necessary to move outside of the given paradigms anddemonstrate how these various approaches can be combined to provide moreexplanatory frames of reference than any one of them can do by itself. In thispaper such an integrative theory is proposed. It preserves the integrity of theconstituent concepts and theories and provides a basis for cross-paradigmstudies. This theory of motivation, volition, and performance (MVP) buildsupon an established integrative theory but expands it by incorporating theconcept of intentions, action control, and information processing within theframework of a system model. Furthermore, the theory illustrates how envi-ronmental (external) influences on behavior combine with internal psycho-logical constructs and processes in relation to goal directed effort, perform-ance, consequences, and outcomes. The paper concludes with a discussion ofimplications of the theory for research and practice.

Keywords: Motivation, intentions, volition, motivational design, ARCS model, self-regulation, performance, learning.

79

____________________________*Corresponding author: Telephone: 850 906-9986; Fax: 850 644-8776; E-mail: [email protected]

Tech., Inst., Cognition and Learning, Vol. 6, pp. 79-104 © 2008 Old City Publishing, Inc.Reprints available directly from the publisher Published by license under the OCP Science imprint,Photocopying permitted by license only a member of the Old City Publishing Group

INTRODUCTION

There are many theories and models that explain aspects of motivation,volition, and learning but most of them tend to stand alone as relativelyindependent areas of inquiry. For example, a well-established theory pertainingto motivation is the ARCS model (Keller, 1999a) which provides a synthesis ofmotivational concepts and theories and a motivational design process. Withregard to volition, Kuhl’s action control theory (Kuhl, 1985), Gollwitzer’s workon implementation intentions (Gollwitzer & Brandstätter, 1997), and research onself-regulated learning behaviors (Corno, 2001; Zimmerman, 2001) are wellestablished in this area. There are many theories of learning, but informationprocessing theory (Atkinson & Schiffrin, 1971; Mayer, 2001) still provides agood foundation for research in the context of learning environment design.And, in addition to these relatively comprehensive theories are the numerousmore specific concepts and theories in each of these areas. This is not a problemfor the scholars who work within specific domains of inquiry and use researchparadigms that are well-established within their respective areas, but it is aproblem for researchers who want to develop more comprehensive and cross-paradigm programs of inquiry, and for designers who are trying to createeffective learning environments that meet the needs of the intended audiences.There are a few theories and models, some of them quite recent (Astleitner &Wiesner, 2004), that attempt to bridge between motivation, volition, andlearning, but each of them has limitations that are addressed in the current paperwhich presents a more comprehensive and integrative theory that can be used forprediction and design.

Thus, the aim of this paper is to present a theory that integrates these andother factors and illustrates how they are connected within a systems perspectivethat preserves the theoretical foundation of each of the components. This type oftheory can be characterized as a concatenated theory (Kaplan, 1964) in contrastto a hierarchical theory. A concatenated theory attempts to assemble explanatorycomponents pertaining to a central phenomenon, in the present case humanperformance, whereas a hierarchical theory is one in which the component lawsand principles are derived from a set of basic principles, as in a hypothetico-deductive theory.

This distinction was also made by Einstein (Kaplan, 1964) who characterizedhierarchical theory as principle-theory which builds from hypotheticalassumptions through logical extensions to validational studies. Theories in theother class, which he calls “constructive theories”, are more empirically-basedand strive to build from simpler schemes to more complex ones. This is the case

80 KELLER

in the present situation in which a central emphasis is on performance, not justlearning, because performance is an observable behavior that is an outcome oflearning and is also influenced by a variety of psychological, sociological,genetic, and environmental influences. The theory to be presented focusesprimarily on the psychological and environmental influences on performance.Genetic influences are reflected through various motivational and abilityvariables and sociological factors are incorporated into the environmentalinfluences. In this regard the theory to be introduced can also be characterized asa macro theory which means, as the name suggests, that it takes a broad view ofthe motivational, volitional, learning, and environmental factors influencingperformance. Given this ultimate focus on performance and the primary goal ofthe theory to integrate motivational and volitional influences with learningprocesses, a convenient title for the theory is the motivation, volition, andperformance (MVP) theory. This theory is represented in a systems model toshow how all of the parts are interrelated and to illustrate relationships that occuras explained and predicted by the theory. Thus, the phrase MVP model refers tothis systems representation while MVP theory refers to the explanatorystructures represented by the components of the theory.

The MVP theory builds on an existing theory of learning motivation andmotivational design which is described in the following section. This motivationtheory is then described in the context of the original version of a comprehensivemodel of motivation, learning, and performance (Keller, 1979, 1983). Next, theoriginal comprehensive model is expanded into the MVP model byincorporating volitional concepts, an information processing model of learning,and a network of linkages among motivation, volition, and informationprocessing components with respect to learning and performance. The paper alsoincludes discussions of environmental influences and design implications inregard to these components together with summaries of three research studiesthat provide empirical support for elements of the theory. The paper concludeswith a discussion of potential benefits of this theory for research and designtogether with implications for research and practice.

MOTIVATION, LEARNING, AND PERFORMANCE

Motivational research includes numerous micro theories and motivationalconstructs as well as at least one theory that synthesizes these various conceptualareas. The motivational theory to be incorporated in this paper is the ARCSmodel of motivational design (Keller, 1979, 1983, 1999a). It is a well established

AN INTEGRATIVE THEORY OF MOTIVATION, VOLITION, AND PERFORMANCE 81

and validated theory (Astleitner & Hufnagl, 2003; Naime-Diffenbach, 1991;Small & Gluck, 1994) that synthesizes motivational theories and constructs intofour categories based on shared attributes. It also provides a systematic approach(Keller, 1999b) to diagnosing motivational problems and designing motivationaltactics into instruction. It incorporates needs assessment based on an analysis ofthe target audience and existing instructional materials, supports the creation ofmotivational objectives and measurements based on the analyses, providesguidance for creating and selecting motivational tactics, and follows a processthat integrates well with instructional design and lesson planning (Keller, 2000February). The analysis of motivational needs and corresponding selection oftactics are based on four dimensions of motivation that were derived from thesynthesis of motivational concepts and theories, and are known as attention (A),relevance (R), confidence (C), and satisfaction (S).

However, the motivational dynamics represented by the four categories of theARCS model do not occur in isolation from other influences on learning andperformance. Their interactive influences can be illustrated in a process, orsystems, model of motivation, learning, and performance (Figure 1). This modelillustrates how a person’s curiosity (“attention”) and motives, or values,(“relevance”) combined with his or her expectancy for success (“confidence”)will determine which goals have the highest saliency and will therefore lead topurposeful effort to accomplish the goal. Environmental influences such asteacher enthusiasm and social values, quality of instruction, clarity ofexpectations, and availability of resources will also influence goal-directedeffort and lead toward some degree of achievement, or performance, which willalso be influenced by one’s knowledge, ability, and skills. One’s performancecombined with the way in which reinforcement contingencies are administereddetermines the consequences of achievement with respect to whether it leads toan expected outcome. These consequences combined with one’s cognitiveevaluations and reflections determine levels of satisfaction with the process andoutcomes. With respect to feedback loops, represented by dotted lines (Figure 1),one’s perceptions of the relationship between effort and performance, andbetween performance and consequences each feed back into and modify one’sexpectancies (Weiner, 1974). Satisfaction level feeds back into motives andvalues to either strengthen or weaken the value one attaches to a given goal.Even though these lines follow a traditional representation as feedback loopswhich is convenient for graphical purposes, this model is a process model thatillustrates how various internal and environmental events influence behaviorover time. Therefore, the feedback lines are in essence feed forward lines. Aperson’s motives and values at time a will result in a chain of events as

82 KELLER

represented in the model. When an outcome is reached and internally,cognitively, evaluated, the results of one’s feelings of satisfaction ordissatisfaction on motives and values would be at time b. Hence, the modelrepresents a series of cycles, not changes in a static situation. However, motivesand expectancies are not always the starting point in the process. It is possible tobegin with a set of observations at any point in the process and examine theirrelationship to other elements of the model.

This macro model has been useful in guiding inquiry on motivation andlearning and providing a basis for identifying design issues. For example,audience analysis can be used to identify sources of motivation and learningproblems and to then design solutions that incorporate the appropriate strategies(environmental stimuli) with regard to motivation, instructional design, andcontingency management. As a process model, this macro model illustrates therelationships among various structures but does not detailed illustrations of theactivities that occur within the major components. For that, one would have tolook into the theories and procedures within the given component. For example,with regard to the expectancy-value part of the model, it is assumed that thegoals with the highest valence will automatically result in action in the form ofeffort to accomplish the desired goals (Pintrich & Schunk, 2002). While this is


FIGURE 1A macro-model of motivation, mearning, and performance.

true, this example also illustrates one of the shortcomings of the model that ledto its expansion. Descriptions of the macro-model and the associated ARCSmodel explain the challenges of sustaining learners’ efforts to accomplish agiven goal despite distractions and competing goals, but the model does notinclude specific concepts or procedures pertaining to this problem. Therefore, itwould be useful to expand the model to explain the internal volitional, or self-regulatory, processes together with external supports that can assist learners inmoving from goal selection to goal-directed actions and persistence. Thefollowing sections of this paper explain how this macro model has beenexpanded to include various volitional factors along with a more detailedrepresentation of learning theory for a more comprehensive integrated theory ofmotivation, volition, and performance (MVP).

VOLITION

Historically, motivation was considered to have two levels. The first is “will”which refers to a person’s desires, wants, or purposes together with a beliefabout whether it is within one’s power to satisfy the desire, or achieve the goal(James, 1890; Pintrich & Schunk, 2002). The second level is the act of using thewill, or “volition,” which refers to a process for converting intentions intoactions. In some cases, the mere saliency of a desire, which is another way ofsaying strong motivation, is sufficient to lead more or less automatically toaction, but often, as William James (1890) pointed out, it is necessary to have aconscious effort supported by determination or extrinsic requirements to convertintentions into action, which is known as volition or self-regulation.

Much of motivation research has focused on understanding what people’sgoals are and why they choose them. For example, the original conceptualizationof “will” as being a combination of desires and beliefs about being able toachieve them is reflected in expectancy-value theory (Vroom, 1964) whichpostulates that behavior potential is a function, assumed to be multiplicative, ofthe perceived importance of a given goal in relation to other goals (value) andone’s subjective probability of being able to achieve the goal (expectancy).While this theory has had a powerful influence in motivational theory and is anintegral part of the model illustrated in Figure 1, its primary contribution is inexplaining how people choose a particular goal or set of goals. It provides anecessary but not always sufficient condition for understanding what impelspeople to action and helps them sustain their initial goal intentions, especiallywhen motivation wanes in the face of competing intentions. In this regard, the

84 KELLER

concepts of action control (Kuhl, 1987), implementation intentions (Gollwitzer,1999), and self-regulation (Zimmerman, 1998) are relevant. All of these pertainto the problem of maintaining goal-oriented behavior and overcomingdiscouragement and attrition, problems that have been experienced especially inself-directed learning environments including distributed learning (for example,distance learning and e-learning), and even classroom courses that put a highlevel of scheduling control into the students’ hands.

Kuhl (1985) defines volition as a mediating factor that “energizes themaintenance and enactment of intended actions” (Kuhl, 1985, p. 90) andtherefore goes beyond motivation. Wolters (1998) commented about howstudents can express sincere desires to accomplish a goal but have a verydifficult time in managing competing goals and distractions that interfere withtheir academic work. Similarly, Pintrich and Garcia (1994) pointed out that theinfluence of volition becomes even more important for college students “who,when you talk to them, are very motivated and concerned about doing well, butoften have a very difficult time enacting their intentions, given all the internaland external distractions they confront in college life” (p. 126f). Theseobservations are, of course, readily apparent to teachers, counselors, therapists,or anyone else who tries to facilitate change in people. In this regard, Kuhl’saction control theory was developed in the context of helping people overcomemaladaptive behaviors in their lives while the work of Zimmerman (2001) andCorno (1993) focuses more on self-regulatory behaviors in students.

Kuhl’s theory postulates six action control strategies that can be employed assoon as a goal (“an action tendency”) achieves the status of a current intention.In other words, commitment to achieving a given goal is a prerequisite toemploying the set of action control strategies, which are:

1. Selective attention: also called the “protective function of volition” (Kuhl,1984, p. 125): it shields the current intention by inhibiting the processing ofinformation about competing action tendencies.

2. Encoding control: facilitates the protective function of volition byselectively encoding those features of incoming stimulus that are related tothe current intention and ignoring irrelevant features.

3. Emotion control: managing emotional states to allow those that support thecurrent intention and suppress those, such as sadness or attraction, in regardto a competing intention that might undermine it.

4. Motivation control: maintaining and reestablishing saliency of the currentintention, especially when the strength of the original tendency was notstrong (“I must do this even though I don’t really want to.”)


5. Environment control: Creating an environment that is free of uncontrollabledistractions and making social commitments, such as telling people whatyou plan to do, that help you protect the current intention.

6. Parsimonious information processing: Knowing when to stop, makingjudgments about how much information is enough and to make decisionsthat maintain active behaviors to support the current intentions.

Kuhl assumes that processes of action control underlie virtually any kind ofactivity and may be called into play to protect a current intention when thestrength of competing tendencies grows too strong or one’s ability to perform theintended actions is weakened (Kuhl, 1985). The effectiveness of employingaction control strategies has been confirmed in many studies in a variety ofbehavior change settings (Kuhl, 1987) as well as in educational settings (Corno,2001; Zimmerman, 2001). Consequently, action control strategies will be animportant component of the MVP theory. However, action control theory doesnot provide detailed explanation of the process by which one moves from thestate of having a predominant goal to active commitment to achieving the goal.For this, Gollwitzer’s work (Gollwitzer, 1996) on intention commitment, orimplementation intentions is helpful.

The first step in moving from desire to action, that is, from the identificationand acceptance of a personal goal to a set of actions to accomplish the goal isthat of intention formation. On the one hand, the concept of “good intentions” isused as a rationalization when things go wrong, or an excuse for not takingaction as in the expression, “the road to hell is paved with good intentions.” But,on the other hand, intentions can be a powerful influence on goalaccomplishment. In a laboratory study with preschool children who were askedto work on a repetitive, boring task that was interrupted with a temptingdistraction (a clown head encouraging children to select and play with toysinstead of working on their assigned task), Patterson and Mischel (1976) testedthe effects of task-facilitating intentions versus temptation-inhibiting intentions.The children were told that a clown box might tempt them to stop working. Thetask-facilitating group was told to keep their attention on the task if thishappened, and the temptation-inhibiting group was told to direct their attentionaway from the clown box. This study and subsequent research (Gollwitzer &Schaal, 2001) shows that temptation-inhibiting intentions have the superioreffect no matter whether motivation to perform the task is high or low. However,these studies have been conducted in laboratory settings and it would beinteresting to see if the same strong effects can be observed with older studentsin learning environments which often contain many distractions.

86 KELLER

Implementation intentions are but one part of a more complete model,called the Rubicon model, that explains that transition from goals to action(Gollwitzer, 1996), This model has some elements in common with Keller’smacro model and others that are not which will be important additions in thedevelopment of the MVP model. The first phase is a pre-decisionalmotivation phase that includes activities associated with identifying andchoosing goals, but stops short of making a full commitment to pursuing agiven goal. It is based on expectancy value theory and is similar to themotivational (values and expectancies) phase in Keller’s macro model (Figure1), but does not have the same level of inclusion of motivation elements thatcan influence goal selection and expectancies. At the conclusion of this phaseone moves forward into the committed stage, which is why it ismetaphorically called the Rubicon model, This refers to Julius Caesar’sfamous decision to cross the Rubicon into Italy proper, which was against thelaw and from which there was no turning back. This “committed phase” is thepre-actional phase of volition which includes committing oneself to actiondeveloping what Gollwitzer calls “strong” intentions that result fromanticipating and immunizing oneself against obstacles. The third phase coversvolitional, or self-regulatory, activities that help sustain goal seekingbehaviors. It is in this part of the Rubicon model that Kuhl’s action controltheory can add depth in regard to volition. The final phase consists of post-actional motivational behaviors such as reflecting on one’s actions andconducting a self-evaluation of outcomes. This is similar to the cognitiveevaluation, equity part of Keller’s macro model (Figure 1).

INTEGRATING VOLITIONAL INFLUENCESINTO THE MVP MODEL

MVP theory expands Keller’s macro model to incorporate action controltheory and implementation planning. First, however, it is important to note that amajor graphical modification was made when designing the MVP model. First,the bottom row of Keller’s original model (Figure 1) containing inputs from theenvironment was moved to the top (Figure 2) and relabeled as “external inputs,”which better communicates the variety of social, cultural, and physicalenvironmental factors that can influence individual motivation and performance.Also, this new placement facilitates communication of the interactive influences ofenvironmental and internal psychological processes on each other and on the majorcategories of outputs, or measurable behaviors, associated with each stage.


With respect to modifying the substance of the macro model, themotivational, or expectancy-value) portions of the model is the same. Gollwitzerhas a “Motivation: Pre-decisional phase” in the Rubicon model, but Keller’soriginal formulation, which includes expectancy theory, is more complete.However, substantial changes were incorporated in the MVP model by adding a

pre-actional stage of volition patterned after Gollwitzer’s (1999) work(Figure 2). This provides a more complete representation of the processes thatoccur in moving from initial motivation to sustained effort. An action stage hasalso been added to include volition as represented in the six components ofKuhl’s action control theory and other concepts of self-regulation (Corno, 2001;Zimmerman, 2001). Figure 2 illustrates the relationships that occur among thesecomponents in the form of internal motivational/volitional processing and alsoillustrates how each of them can be related to an observable outcome variable.However, the magnitudes of the relationships are not illustrated in the diagram.For example, if ‘effort direction’ is extremely high, then it might not benecessary for the learner to invoke implementation intentions or action controlstrategies. This condition would be similar to a state of flow (Csikszentmihalyi,1990) in which the learner is fully absorbed in the immediate activity and notrelying on metacognitively induced strategies of self-regulation to stay on task.The volitional elements would become activated when necessary to protect theintention if it began to wane for the previously described reasons (too manycompeting tendencies or excessive challenge in goal pursuit). This conditionwould be similar to a state of flow (Csikszentmihalyi, 1990) in which the learneris fully absorbed in the immediate activity and not relying on metacognitivelyinduced strategies of self-regulation to stay on task.

In keeping with these two new elements of the model, two outputs wereadded to clarify the outcomes of motivation, pre-actional planning, and actioncontrol. Instead of the unitary designation of effort (Figure 1) the result ofchoosing a goal is called effort direction (Figure 2), which refers to a person’sdesires, or goal choices. Next is effort initiation which refers to the initiation ofeffort to achieve the chosen goals. The third is effort persistence which isinfluenced by the strength of the goal, the level of commitment, and the use ofvolitional strategies if any are employed by the learner. These can be measuredin many ways ranging from self-reports through direct observation toperformance indicators such as on-task versus procrastination times.

The remaining part of the Rubicon model, “Motivation: Post Actional Phase,”is similar to the cognitive evaluation component of Keller’s original model(Figure 1), but has some useful distinctions related to reflection and emotionalprocessing. It is now called “outcomes processing” (Figure 2) and includes these

88 KELLER

elements. Also, note that the influence of behavior management by means ofreinforcement contingencies has been retained from Keller’s original model, butthere is a new block (Figure 2) within external inputs that refers to strategies thatcan be implemented to influence how people manage their emotional andevaluative reactions to events.

In summary, keep in mind that the goal of the MVP, as with Keller’s originalmacro model and the ARCS model, is to provide an explanatory shell thatsynthesizes theories and models based on shared attributes, and organizes themby means of a general systems model that illustrates mutual influences on otheraspects of learning and performance. This is not a glib aggregation based on afew similarities in semantics or “surface level” attributes, but is based on thetheory, purpose, and empirical research underlying each of these components.However, as synthesis requires generalization, some of the unique properties ofeach of the component theories and constructs will not be highlighted in thisconcatenated theory. This theory does not purport to supplant any of thecomponent theories and constructs, but only to portray their relationships withthe aim of facilitating new directions in research and aiding designers indiagnosing and developing solutions to human performance problems.


Goals,

Desires

Motivation & Volitional Processing

Interest

Curiosity

MotivesValues

Expect-ancies

Volitional

strategies

influencing

support of

self-regulatory

actions

Learning &Performance

Effortdirection

Pre-Action

Planning

Emotions

(Positive

&Negative)

Commit-

mentIntentions

Actions

Action

controlstrategies

(Self-reg-

ulatory

actions)

EffortPersistence

Motivational

strategies

influencing

curiosity,

values, &

expectancies

Effort

initiation

Implementa-

tion strategies

influencingpre-action

planning

Influence &interact with:

Resulting in:

External

Inputs

Psycho-

logicalEnviron-

ment

Outputs

Abilities,

knowledge,

skills

Consequences

Satisfaction

Reflections & Emo-tional Reactions

including

Cognitive evaluationand perceived

equity

Contingency

management

and intrinsicreinforcement

strategies

Outcomes Processing

Strategies

influencing

cognitive &

emotional

outcome

evaluations

FIGURE 2Integration of volitional and post-actional concepts into Keller’s macro model.

INFORMATION PROCESSING

Another limitation of Keller’s (1983) macro model is that it does not illustratehow information processing elements are part of the learning process and howthey interact with motivation and volition. In other words, even though Keller’smacro-model illustrates that abilities, knowledge, and skills influenceperformance (Figure 1), it does not illustrate the dynamics of perceptual andcognitive activities involved in learning. Therefore, the next step in expandingthe macro model was to replace the simple designation of knowledge, ability,and skill with an information processing model (Figure 3) based on Atkinson &Shifferin’s original formulation (1971), but modified to incorporate distinctionsbetween words versus pictures or, more precisely, digital versus iconic signs(Knowlton, 1966; Langer, 1942) as environmental inputs because of thedifferences in the ways they can affect perception, attention, motivation, andlearning. Paivio (Paivio, 1969, 1971) explained this by postulating dualprocessing theory which assumes that there are two different encoding systemsin working memory for perceiving and processing digital (verbal, textual) versusiconic (pictures, diagrams) information. Many of these concepts have beenincorporated into research on learning in a multimedia environment which hasre-emerged as a major area of interest given the breadth and depth of technologyintegration into instruction. For example, Mayer’s (2001) research on learningfrom pictures versus words reconfirms and adds to the large volume of researchon this topic that was done in the 1960s and 1970s (Levie & Dickie, 1973; Levie& Lentz, 1982).

Another important concept to include in conjunction with general informationprocessing theory is cognitive load (Paas, Renkl, & Sweller, 2003), which refersto the amount of information that a person can process in working memory. Thisis another of the problematic situations in many learning environments,especially when learning complex cognitive skills or working with multimediabecause there are several additions to cognitive load that are not typicallypresent in a more structured lecture or textbook situation. For example, Sweller(1994) distinguishes between intrinsic cognitive load which refers to theinherent complexity of the material to be learned, germane cognitive load whichrefers to the level of cognitive activity resulting from task-oriented efforts of thelearner, and extraneous cognitive load which is induced by instructionalactivities and distractions that are divergent and not effective in accomplishingthe task. In traditional self-directed learning environments, such as programmedinstruction, it was difficult to manage intrinsic cognitive load. Programs ofinstruction would be used by students with differing entry levels of knowledge

90 KELLER

and skill, but it was not feasible to provide a single program that would notproduce cognitive overload for some students and under-load for others despitethe efforts to include features such as branching. In multimedia instruction it iseven more challenging because of the number and types of stimuli that can occurin this environment. With the availability of hot links, internet excursions, multiplestimuli (rollovers, branching topics, navigational innovations) within themultimedia program, the potential for excessive cognitive load is high, especially inthe area of extraneous cognitive load. Deimann & Keller (2006) documented manyof these problems such as “lost in hyperspace” and the “serendipity effect” and howthey can be managed by the appropriate use of volitional strategies. Cognitive loadproblems can also occur when learning complex cognitive skills. The intrinsic loadtends to be very high, which means that extraneous load in the form of competingintentions or distractions can lead to excessive cognitive load and the need to invokeaction control strategies to stay on task.

Information processing theory combined with motivation and volitionprinciples has implications for principles of cognitive load and how to bestdesign instruction to maximize germane, or useful, cognitive load (Sweller,1994) while minimizing extraneous, or dysfunctional, cognitive load. One canfind support for these assumptions in the general principles of message design


FIGURE 3An information processing model incorporating dual processing theory.

compiled by Fleming & Levie (1978), as well as many other establishedprinciples of perception and memory that could apply to and be empiricallytested in the design of instruction. One such example, based on the work ofAnderson and Faust (1973) and which could be presumed to be related tocognitive load is Principle 2.12 (Fleming & Levie, 1978) “Learning is facilitatedwhere criterial cues are salient (dominant, apparent, conspicuous). Add non-criterial cues only if and as necessary” (p. 115). Unnecessary cues would add toextraneous cognitive load. Additional implications are discussed below in thedescriptions of the final restructuring of the original macro model.

MOTIVATION IN RELATION TO INFORMATION PROCESSING

One of the limitations of information processing theory in general, includingMayer’s, is that they do not include motivational or volitional considerations.Astleitner & Wiesner (2004), building on the work of Mayer (2001), Rheinberg,Vollmeyer, and Rollet (2000), and Keller (1983, 1997, 1999a) have proposed anintegrated theory of information processing and motivation that includes“motivational processing” as well as elements of mental resource management(Figure 4). They draw primarily on Kuhl (1985) who postulated that such thingsas wishes, intentions, values, and emotions are also part of working memory inaddition to the traditionally recognized perceptual and cognitive processingcomponents. The motivational components specified by Astleitner & Wiesner(Figure 4) are goal setting and action control. They are connected to theinformation processing model by means of mental resource managementactivites such as attention, engagement, and monitoring (Figure 4). It should benoted that their use of the term “attention” refers to actions that facilitatelearning, such as providing cues to focus attention to salient parts of the mentaltasks at hand. This is different from using “attention” in the motivational sense(Keller, 1999a) which refers to stimulating and sustaining arousal and curiosity.Astleitner & Wiesner’s model has promise for assisting in conceptualizingresearch on the interplay of cognitive and motivation elements in learning, asillustrated by the authors’ presentation of a list of general predictions, especiallylearning in a multimedia environment which is the focus of their paper.However, their model has limitations that can be illustrated by the third and finalexpansion of Keller’s macro model which incorporates both informationprocessing theory with dual encoding specifications and a revised version ofAstleitner & Wiesner’s model together with several integrating elements that arenot present in either of those models.

92 KELLER

ADDING INFORMATION PROCESSING ANDINTERNALLINKAGES TO THE MVP MODEL

The final phase of modifications to Keller’s macro model (Figure 1) resultsfrom adding the information processing components into the fully expandedmodel (Figure 5), specifying cognitive load as a component of workingmemory, adding a psychomotor component represented by “practice,” andillustrating their linkages with motivational and volitional components. In theinformation processing elements in Figure 5 the dual processing elements arenot distinguished due to the already complex structure of the model, but theyare presumed to be within the sensory inputs and working memorycomponents. Although there are numerous control processes within workingmemory (Atkinson & Schiffrin, 1971), attention is called to the concept ofcognitive load (Sweller, 1994) in the present model. This variable is presumedby many to be a key factor in designing instruction, especially multimedia,where the stimulus arrays can be complex and distracting and in designinginstruction for the teaching of complex cognitive skills (Van Merriënboer,Kirschner, & Kester, 2003). And, the concept of practice is included because,as shown by Ericsson (2006), the type of practice that distinguishes superior


FIGURE 4A model of multimedia, learning, and motivation (Astleitner & Wiesner, 2004).

performers, called deliberate practice, is a behavior that combinespsychomotor, cognitive, and motivational elements in the development andmaintenance of expert performance.

Astleitner and Wiesner ’s modification of Mayer ’s model has beenincorporated in two ways. The first is to include “mental activities” in theinformation processing section just as they did. But, “mental resourcemanagement” has been expanded in terms of its elements and placed into aseparate category called “Motivation & Information Processing Interface.”This makes it easier to illustrate the direct interactions between the two areassuch as the effect of higher levels of motivation on higher capacity forgermane cognitive load and how goals and implementation intentions caninfluence focus of attention which influences mental activities such asselection and organization of stimuli.

With respect to the remaining part of Astleitner and Wiesner’s model,called “Motivational Processing,” the component parts of this section, whichare goal setting and action control, are already part of the MVP model in thesection called “Motivation and Volitional Processing.” Goal setting iscontained within the pre-decisional phase of motivation, and action control isin the actional phase of volition.

94 KELLER

FIGURE 5The MVP model of motivation, volition, and performance.

SUMMARY OF MODIFICATIONS TO &RESTRUCTURING OF THE MODEL

By expanding the original model of motivation, learning, and performance(Keller, 1979, 1983) to include implementation intentions and action control andby means of other modifications that have been described, the proposed newmacro-model seems to offer a more comprehensive basis for developing theoryand research on the dynamics of motivation, volition, and informationprocessing in regard to learning, performance, and instructional design.

IMPLICATIONS OF THIS MODEL FOR RESEARCHON CONCEPTS AND THEORY

What, one may ask, is the value of this type of theoretical integration? High-level theories can be confusing and are by nature complex. Furthermore, forevery class of variables and specification of relationships that are included, thereare others that could be included. Models such as these have the danger ofbecoming like an expanding universe as each version leads to the incorporationand conceptualization of multiplicatively more concepts, subordinate models,and relationships. But, even at this level of theorizing, it is important to maintaina sense of parsimony and testability. In the present situation, the goal is tocontinue efforts to explain the dynamics and inner workings of motivation,learning, achievement, and satisfaction as it results in continuing motivation.Therefore, the model has been expanded only insofar as necessary to integrate afew major concepts and variables that will help accomplish this goal.

An area of research that can be used as an example is curiosity which, as apsychological construct, has a long history of investigation and was initiallygrounded in conceptualizations of deprivation and sensation-seeking tendencies(Berlyne, 1954). A more formal conceptualization was provided by Berlyne(1954) who expanded this concept by making a distinction between perceptualcuriosity which refers to one’s more or less automatic response to novel andincongruous stimuli and epistemic curiosity referring to a more intellectual levelof curiosity in which one seeks answers to complex and incongruous situations.He also introduced a distinction between specific and diversive curiosity whichwas elaborated by Day (1968). Specific curiosity is aroused by a situation ofinterest that has the properties of complexity, novelty, and ambiguity. Thespecifically curious person tends to focus on the task with the aim of reducinguncertainty by means of exploration and gaining information. In contrast,


diversive curiosity is characterized by tendencies to be easily distracted bynew stimuli, lack of follow though on tasks, and frequently becoming boredand seeking change. Research on curiosity has focused on refinements ofthe construct, developing measures, and studying conditions that stimulateone type of curiosity or another (Collins, Litman, & Spielberger, 2004; Day,1968; Litman & Spielberger, 2003; Maw & Maw, 1977). But, it hasremained as a somewhat isolated area of inquiry as is illustrated by itsplacement in the motivation portion of the MVP model, and even recentstudies tend to focus on further definitions of the construct and itsrelationship to other constructs such as flow and sensation seeking(Kashdan, Rose, & Fincham, 2004). However, a perusal of the MVP modelcan stimulate the thought that it might be fruitful to consider curiosityconcepts such as specific versus diversive curiosity in conjunction withvolitional theory and learning environment design. It would seem thatcuriosity would be positively correlated with achievement in a stimulus-richenvironment such as a Montessori classroom or online multimediaenvironment which typically have characteristics that stimulate exploratorybehavior. However, these environments also have the potential for beingoverwhelming because of their many opportunities for exploration. This couldbe especially true when one considers individual differences such as specificversus diversive curiosity in relation to the conditions that are optimal for eachcuriosity type, and how the introduction of volitional supports might differ foreach group. For example, people with specific curiosity could be expected torequire less volitional support than those with diversive curiosity in a stimulusrich environment. In summary, becoming familiar with the variety of areas ofresearch that are synthesized into the MVP model, one can be in a better positionto envision new lines of research.

Another example can be illustrated by the concept of computer rage(Seligman, 2005) which appears to be occurring with greater and greaterfrequency and has become a topic of interest in computer interface design(Klein, Moon, & Picard, 2002; Selvidge, Chaparro, & Bender, 2002). Likemany new areas of inquiry, this one began as an investigation of a particularphenomenon which then gradually expanded to incorporate theory. Ceaparuet al (2004) conducted a series of self-report and observational measures attwo universities to determine causes and severity of user frustration whenusing computers. The primary components that they identified are errors, timedelays, and emotional reactions. Their study also provided empiricalverification of similar observations reported by others (Klein, Moon, &Picard, 2002; Selvidge, Chaparro, & Bender, 2002).

96 KELLER

In addition to this research on documenting causes of problems, researchersare endeavoring to identify solution areas and theoretical bases forunderstanding the phenomenon. The most common explanation is grounded inthe traditional frustration-aggression hypothesis (Dollard et al., 1939). However,this provides only a partial explanation because even though it has beenconfirmed consistently in many studies that aggression is caused by frustration(Berkowitz, 1989), the converse is not true; that is, frustration does not alwayslead to aggression. Consequently, researchers have been focusing on ways ofalleviating frustration without aggression. The primary methods include removalof the causes of frustration by improving the quality and reliability of computerinterfaces and by providing frustration alleviating feedback in the form ofapologies and other kinds of messages (Klein, Moon, & Picard, 2002). Toprovide theoretical support for this approach, researchers are integrating theoriesof human emotion and emotional management, even to the point of havingcomputers be sensitive to emotional changes in the user and then providing anappropriate response (Klein, Moon, & Picard, 2002; Scheirer, Fernandez, Klein,& Picard, 2002).

Even though this area of inquiry is broadening, it could perhaps expand evenmore quickly into relevant and potentially useful areas by examining anintegrative theory such as that represented in the MVP model. For example, ifwe try to give an overall characterization of this research with reference to theMVP model, we can see that user efforts to perform well (refer to the “outputs”row, Figure 7) are frustrated by the environmental conditions (external inputs)that either facilitate or restrict performance due to poor man-machine interfaces.This results in emotional frustration (outcomes processing) and a frequenttendency to behave aggressively toward the equipment. The concept offrustration-aggression also falls into the outcomes processing “box” (Figure 7)as an emotional reaction. However, not everyone responds the same way tofrustration or to interventions such as apologies from the computer, which somepeople regard as inappropriate “anthropomorphizing” of the machine. It mightbe fruitful to incorporate investigations into goal setting (the stronger one’s goalmotivation, presumably the greater one’s frustration in being thwarted fromachieving the goal), volition to better understand how to immunize users againstthese inevitable frustrating obstacles in computer environments, and theinfluence of interface design on ease of information processing and cognitiveload in order to preserve as much working memory as possible for the task-at-hand. It is, of course, possible and even perhaps likely that researchers wouldeventually identify some of these avenues of inquiry, but the contention of thispaper is that an integrative model can facilitate and speed-up the process, and


even perhaps stimulate ideas that would not have otherwise occurred to people.Implications of This Model for Research on Practice

With respect to practice, there are numerous ways in which the MVP modelcan be applied. First, it is important to note that there are some fundamentaldifferences between the use of theory in research and practice. For example, agoal of research is to find general principles and in order to do so, the typicalparadigm of science is to vary the parameters of interest in a study and holdother potentially confounding parameters constant. However, this is not possiblein practice. Solutions to problems or the introduction of innovations is done inactual settings where it is not possible to hold major components of the systemconstant. It is common in practice to have innovations or solutions, even whenbased on scientifically proven principles, fail or have only limited successbecause of unintended outcomes or uncontrollable obstacles.

Because of these multiple levels of interaction, system theory provides aneffective approach to representing the phenomena under investigation. Bothresearch and practice incorporate principles of system thinking. As Whitehead(1938) pointed out, “Connectedness is of the essence of all things of all types”(p. 13). He also said that, “It follows that in every consideration of a single factthere is the suppressed presupposition of the environmental coordinationrequisite for its existence” p. 13.) In any controlled study, there arepresuppositions about conditions that can be held constant which, in the “realworld” are not constant. This is commonly illustrated by many of the principlesin Newtonian physics that presumed a frictionless universe. But it can also beillustrated by contemporary events such as unexpected side effects of medicinesresulting from presuppositions that might not even have been evident during thedevelopment and initial testing of the drug. Even though both research andpractice use system thinking, and both are concerned about inputs and outputs,the emphasis is different. In research the emphasis is more on identifying andcontrolling inputs (independent variables) while in practice the emphasis is moreon anticipating and managing the effects of outputs. For example, with regard topractice, students may achieve mastery in job skill workshops but follow-upevaluations frequently show that the workers are not using their newly acquiredskills. This could be for many reasons including lack of opportunity, lack ofessential resources, conflicting demands on their time, or conflicting incentives.The lesson from this is that in practice, one must take a system-wide perspectiveand attempt to manage the effects of an intervention in one system with respectto its interactive influences with coordinate systems, subsystems, andsuprasystems. When workers are expected to change their skills, then jobdefinitions, incentives, resource supports, and even management procedures

98 KELLER

might have to be changed to accommodate and support the new skill sets. In this regard, each of the major components of the MVP model can be

viewed as a subsystem with multiple interactions with the other subsystems. It iscommon to introduce new instructional strategies or models of instruction into acourse and they tend to be successful as long as they are being implemented bythe advocate who created them, but they tend to die out when the advocateleaves or not succeed when other people try to use the approach. This can bebecause people other than the advocate understand the new approach only as anew instructional technique, which places it within the performance facilitatingexternal inputs portion of the MVP. They might not consider changes that haveto be made in techniques to motivate students, changes in volitionalrequirements, and scaffolding requirements to manage cognitive load. Why,then, one might ask, did the innovation succeed in the hands of the advocate, butnot with others? The advocate might very well have been making theadjustments to the other subsystems, but without being consciously aware of itand not seeing these as essential parts of the intervention which he perceived tobe strictly a new instructional strategy without a need to consider motivational,volitional, or information processing components. The same thing can happen inresearch (Rosenthal & Rosnow, 1969) where researchers are not able toreproduce the findings of the original studies but again, this tends to result fromunintended influences of inputs rather than consequences of outputs.

Thus, the MVP model can be used as a tool for diagnosis and design. Whenthere are problems in a learning system, the MVP model can guide a needsassessment to determine where the sources of the problem are located. Forexample, if students are not scoring well on a standardized test of mathematics,one can examine each input into this performance setting to identify potentialcauses. The problem could be due to a lack of ability on the part of the students,lack of volitional skills, lack of motivation due to boredom, irrelevance of thetask, or lack of confidence, or cognitive overload. It could also be due toineffective design in the learning environment, which falls under external inputs,or a lack of reinforcing consequences when the students do succeed. The benefitof the MVP model is that it facilitates a systematic examination of all thesefactors and then provides the basis for a coordinated set of improvements thattake all the relevant factors into consideration. Its utility can also be expanded byintegrating it with instructional design models such as the 4C/ID-model (VanMerriënboer, Jelsma, & Paas, 1992) which includes four major components that arepresumed to be essential in building blueprints for complex learning, especially inregard to the components called “supportive information” and “just-in-timeinformation.” By integrating motivational and volitional support systematically


with instructional support, there is a much greater likelihood that students willmaintain their persistence to succeed when trying to master the learning of complexcognitive skills. This type of coordinated process was demonstrated (Bickford,1989) with the ARCS model in which audience analysis identifies what the specificmotivational deficits are in a given situation, and then motivational strategies wereprepared and integrated with the instructional strategies that resulted from theapplication of a systematic instructional design process.

CONCLUSION

The overall purpose of this paper was to present process-oriented modelgrounded in a system representation of a set of logical interfaces betweenseveral major theories and models pertaining to motivation, volition, learning,and performance. The focus has been on building a frame of reference forincorporating existing research and practices in an organized manner thatfacilitates efforts to move outside the boundaries of defined areas of researchin order to study the relationships between and among these areas. Keller’smacro-model of motivation, learning, and performance (1983) was retained toprovide the initial structural and process frames of reference. Majormodifications to the model consisted of the incorporation of the concepts ofimplementation intentions (Gollwitzer & Brandstätter, 1997) and actioncontrol (Kuhl, 1987). Both of these have their own history of theory andresearch and provide useful additions to the model. They also help provide abasis for studying and explaining the relationships between motivation andmental activities in working memory. To facilitate this set of understandingsand relationships, Astleitner and Wiesner’s (2004) integrated model ofmultimedia learning and motivation was incorporated into the presentstructure, but not without modification. They proposed the addition of sets ofactivities for mental resource management and motivational processing toMayer’s (2001) multimedia learning theory. In the present model, themotivational processing elements were incorporated in the motivational andvolitional components of the MVP model, and they were supplemented byother motivational elements that help constitute a more holistic considerationof motivational factors. The result of this process is a frame of reference, asconfirmed by early empirical studies, that has the potential for offering amore comprehensive explanation of the relationships among motivation,learning, and performance and supports continued theory building, empiricalstudies, and application strategy development and testing.

100 KELLER

ACKNOWLEDGEMENT

I am grateful for the contributions of Dr. Markus Deimann, Fern Universitätin Hagen, Germany, to the expansions of many of my ideas about motivationwhile he studied with me here in Tallahassee as a visiting scholar. Several of theseminal ideas that led to this paper were stimulated by our work together.

REFERENCES

Anderson, R. C., & Faust, G. W. (1973). Educational psychology: The science of instruction andlearning. New York: Dodd, Mead & Co.

Astleitner, H., & Hufnagl, M. (2003). The effects of situation-outcome-expectancies and of ARCS-strategies on self-regulated learning with web-lectures. Journal of Educational Multimedia andHypermedia, 12(4), 361-376.

Astleitner, H., & Wiesner, C. (2004). An integrated model of multimedia learning and motivation.Journal of Educational Multimedia and Hypermedia, 13(1), 3-21.

Atkinson, R. C., & Schiffrin, R. M. (1971). The control of short-term memory. Scientific American,225, 82-90.

Berkowitz, L. (1989). Frustration-aggression hypothesis: examination and reformulation.Psychological Bulletin, 106(1), 59-73.

Berlyne, D. E. (1954). An experimental study of human curiosity. Br J Psychol, 45(4), 256-265.

Bickford, N. L. (1989). The systematic application of principles of motivation to the design ofprinted instructional materials. Unpublished Dissertation, Florida State University,Tallahassee.

Ceaparu, I., Lazar, J., Bessiere, K., Robinson, J., & Shneiderman, B. (2004). Determining causes andseverity of end-user frustration. International Journal of Human-Computer Interaction, 17(3),333-356.

Collins, R. P., Litman, J. A., & Spielberger, C. D. (2004). The measurement of perceptual curiosity.Personality and Individual Differences, 36, 1127-1141.

Corno, L. (2001). Volitional Aspects of Self-Regulated Learning. In B. J. Zimmerman & D. H.Schunk (Eds.), Self-Regulated Learning and Academic Achievement. Theoretical Perspectives(Second Edition) (pp. 191-226). Mahawah, N.J.: Erlbaum.

Corno, L., & Kanfer, R. (1993). The role of volition in learning and performance. Review of Researchin Education, 19, 301-341.

Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper & Row.

Day, H. I. (1968). Role of specific curiosity in school achievement. J Educ Psychol, 59(1), 37-43.

Deimann, M., & Keller, J. M. (2006). Volitional aspects of multimedia learning. Journal ofEducational Multimedia and Hypermedia, 15(2), 137-158.

Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., Sears, R. R., & Yale University. Institute ofHuman Relations. (1939). Frustration and aggression. New Haven,: For the Institute of HumanRelations by Yale University Press.


Ericsson, K. A. (2006). The influence of experience and deliberate practice on the development ofsuperior expert performance. In K. A. Ericsson, N. Charness, P. Feltovich & R. R. Hoffman(Eds.), Cambridge handbook of expertise and expert performance (pp. 39-68). Cambridge,UK: Cambridge University Press.

Fleming, M., & Levie, W. H. (1978). Instructional message design: Principles from the behavioralsciences. Englewood Cliffs, NJ: Educational Technology Publications.

Gollwitzer, P. M. (1996). The Volitional Benefits of Planning. In P. M. Gollwitzer & J. A. Bargh(Eds.), The Psychology of Action. Linking Cognition and Motivation to Bahavior (pp. 287-312). New York: Guilford Press.

Gollwitzer, P. M. (1999). Implementation Intentions. Strong Effects of Simple Plans. AmericanPsychologist, 54(7), 493-503.

Gollwitzer, P. M., & Brandstätter, V. (1997). Implementation Intentions and Effective Goal Pursuit.Journal of Personality and Social Psychology, 73(1), 186-199.

Gollwitzer, P. M., & Schaal, B. (2001). How Goals and Plans Affect Action. In J. M. Collis & S.Messick (Eds.), Intelligence and Personality. Bridging the Gap in Theory and Measurement(pp. 139-161). Mahwah, N.J.: Erlbaum.

James, W. (1890). The principles of psychology (Vol. 2). New York: Henry Holt.

Kaplan, A. (1964). The conduct of inquiry. San Francisco: Chaldler.

Kashdan, T. B., Rose, P., & Fincham, F. D. (2004). Curiosity and exploration: facilitating positivesubjective experiences and personal growth opportunities. J Pers Assess, 82(3), 291-305.

Keller, J. M. (1979). Motivation and instructional design: A theoretical perspective. Journal ofInstructional Development, 2(4), 26 - 34.

Keller, J. M. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.), Instructionaldesign theories and models: An overview of their current status. Hillsdale, NJ: LawrenceErlbaum Associates.

Keller, J. M. (1997). Motivational design and multimedia: Beyond the novelty effect. StrategicHuman Resource Development Review, 1(1), 188-203.

Keller, J. M. (1999a). Motivation in cyber learning environments. Educational TechnologyInternational, 1(1), 7 - 30.

Keller, J. M. (1999b). Using the ARCS Motivational Process in Computer-Based Instruction andDistance Education. New Directions for Teaching and Learning, 78, 39-48.

Keller, J. M. (2000 February). How to integrate learner motivation planning into lesson planning:The ARCS model approach. Paper presented at the Paper presented at VII Semanario, Santiago,Cuba.

Klein, J., Moon, Y., & Picard, R. W. (2002). This computer responds to user frustration: Theory,design, and results. Interacting with Computers, 14, 119-140.

Knowlton, J. Q. (1966). On the definition of a picture. AV Communication Review, 14(157-183).

Kuhl, J. (1985). Volitional mediators of cognitive-behavior-consistency; self-regulatory processesand action versus state orientation. In J. Kuhl & J. Beckmann (Eds.), Action control. Fromcognition to behavior (pp. 101-128). Berlin: Springer.

Kuhl, J. (1987). Action control: The maintenance of motivational states. In F. Halisch & J. Kuhl(Eds.), Motivation, Intention and Volition (pp. 279-291). Berlin: Springer.

Langer, S. K. (1942). Philosophy in a new key: A study in the symbolism of reason, rite, and art.Cambridge: Harvard University Press.

102 KELLER

Levie, W. H., & Dickie, K. E. (1973). The analysis and application of media. In R. M. W. Travers(Ed.), Second Handbook of Research on Teaching: The American Educationl ResearchAssociation.

Levie, W. H., & Lentz, R. (1982). Effects of text illustrations: A review of research. EducatonalCommunication and Technology Journal, 30(4), 195-232.

Litman, J. A., & Spielberger, C. D. (2003). Measuring epistemic curiosity and its diversive andspecific components. Journal of Personality and Assessment, 80, 75-84.

Maw, W. H., & Maw, E. W. (1977). Nature and assessment of human curiosity. In P. McReynolds(Ed.), Advances in psychological assessment (Vol. 4). San Francisco: Jossey-Bass.

Mayer, R. E. (2001). Multimedia Learning. Cambridge: Cambridge University Press.

Naime-Diffenbach, B. (1991). Validation of attention and confidence as independent components ofthe ARCS motivational model. Unpublished Ph.D., Florida State University, Tallahassee, FL.

Paas, F., Renkl, A., & Sweller, J. (2003). Cognitive Load Theory and Instructional Design: RecentDevelopments. Educational Psychologist, 38(1), 1-4.

Paivio, A. (1969). Mental imagery in associative learning and memory. Psychological Review, 76(3),241-263.

Paivio, A. (1971). Imagery and verbal processes. New York: Holt, Rinehart, and Winston.

Patterson, C. J., & Mischel, W. (1976). Effects of tempation-inhibiting and task-facilitating plans ofself-control. Journal of Personality and Social Psychology, 33, 209-217.

Pintrich, P. R., & Garcia, T. (1994). Self-regulated learning in college students: Knowledge,strategies, and motivation. In C. E. Weinstein (Ed.), Student motivation, cognition, andlearning. Hillsdale, NJ: Erlbaum.

Pintrich, P. R., & Schunk, D. H. (2002). Motivation in Education. Theory, Research, andApplications (Vol. 2). Upper Saddle River, N.J.: Merrill Prentice Hall.

Rheinberg, F., Vollmeyer, R., & Rollett, W. (2000). Motivation and action in self-regulated learning.In M. Boekaerts, P. R. Pintrich & M. Zeidner (Eds.), Handbook of Self-Regulation (pp. 503-529). San Diego: Academic Press.

Rosenthal, R., & Rosnow, R. L. (1969). Artifact in behavioral research. New York,: Academic Press.

Scheirer, J., Fernandez, R., Klein, J., & Picard, R. W. (2002). Frustrating the user on purpose: A steptoward building an affective computer. Interacting with Computers, 14, 93-118.

Seligman, K. (2005, April 21, 2005). Call the geeks when computer rage boils over. San FranciscoChronicle.

Selvidge, P. R., Chaparro, B. S., & Bender, G. T. (2002). The world wide wait: Effects of delays onuser performance. International Journal of Industrial Ergonomics, 29, 15-20.

Small, R. V., & Gluck, M. (1994). The relationship of motivational conditions to effectiveinstructional attributes: A magnitude scaling approach. Educational Technology, 34(8), 33 - 40.

Sweller, J. (1994). Cognitive Load Theory, Learning Difficulty, and Instructional Design. Learningand Instruction, 4, 295-312.

Van Merriënboer, J. J. G., Jelsma, O., & Paas, F. (1992). Training for reflective expertise: A four-component instructional design model for training complex cognitive skills. ETR&D-Educational Technology Research and Development, 40(2), 23-43.

Van Merriënboer, J. J. G., Kirschner, P. A., & Kester, L. (2003). Taking the Load Off a Learner’sMind: Instructional Design for Complex Learning. Educational Psychologist, 38(1), 5-13.


Vroom, V. H. (1964). Work and motivation. New York: John Wiley and Sons.

Weiner, B. (Ed.). (1974). Achievement motivation and attribution theory. Morristown, NJ: GeneralLearning Press.

Whitehead, A. N. (1938). Modes of Thought. New York: Macmillan Company.

Wolters, C. (1998). Self-regulated learning and college student’s regulation of motivation. Journal ofEducational Psychology, 90, 224-235.

Zimmerman, B. J. (1998). Academic Studying and the Development of Personal Skill: A Self-Regulatory Perspecitve. Educational Psychologist, 33(2/3), 73-86.

Zimmerman, B. J. (2001). Theories of Self-Regulated Learning and Academic Achievement: AnOverview and Analysis. In B. J. Zimmerman & D. H. Schunk (Eds.), Self-Regulated Learningand Academic Achievement. Theoretical Perspectives (pp. 1-38). Mahwah, N.J.: Erlbaum.

104 KELLER

An Integrative Theory of Motivation, Volition, and...

Documents

Transcript of An Integrative Theory of Motivation, Volition, and...